Although there is a growing body of research on the use of performance feedback to improve treatment integrity, antecedent interventions have received little attention. Antecedent interventions are simple to use and have a strong research base. This study examined the effects of offering teachers choice of treatment components. A choice condition was compared with a no-choice condition in which an expert-derived treatment was used. An A/B/C/A/C/B design with counterbalancing was combined with a simultaneous-treatments design to examine treatment integrity and student problem behavior for 3 teacher-student dyads. The results indicated that, although both treatments improved student behavior, higher levels of treatment integrity and better behavioral outcomes were associated with the choice condition. The results are discussed in terms of the merits of simple antecedent strategies for improving treatment integrity and the benefit of adding experimental design elements that provide a more direct measure of teacher preference.